Abstract
Purpose
The purpose of this article is the effect of doping minor Ni on the microstructure evolution of a Sn-xNi (x = 0, 0.05 and 0.1 wt.%)/Ni (Poly-crystal/Single-crystal abbreviated as PC Ni/SC Ni) solder joint during reflow and aging treatment. Results showed that the intermetallic compounds (IMCs) of the interfacial layer of Sn-xNi/PC Ni joints were Ni3Sn4 phase, while the IMCs of Sn-xNi/SC Ni joints were NiSn4 phase. After the reflow process and thermal aging of different joints, the growth behavior of interfacial layer was different due to the different mechanism of element diffusion of the two substrates. The PC Ni substrate mainly provided Ni atoms through grain boundary diffusion. The Ni3Sn4 phase of the Sn0.05Ni/PC Ni joint was finer, and the diffusion flux of Sn and Ni elements increased, so the Ni3Sn4 layer of this joint was the thickest. The SC Ni substrate mainly provided Ni atoms through the lattice diffusion. The Sn0.1Ni/SC Ni joint increases the number of Ni atoms at the interface due to the doping of 0.1Ni (wt.%) elements, so the joint had the thickest NiSn4 layer.
Design/methodology/approach
The effects of doping minor Ni on the microstructure evolution of an Sn-xNi (x = 0, 0.05 and 0.1 Wt.%)/Ni (Poly-crystal/Single-crystal abbreviated as PC Ni/SC Ni) solder joint during reflow and aging treatment was investigated in this study.
Findings
Results showed that the intermetallic compounds (IMCs) of the interfacial layer of Sn-xNi/PC Ni joints were Ni3Sn4 phase, while the IMCs of Sn-xNi/SC Ni joints were NiSn4 phase. After the reflow process and thermal aging of different joints, the growth behavior of the interfacial layer was different due to the different mechanisms of element diffusion of the two substrates.
Originality/value
In this study, the effect of doping Ni on the growth and formation mechanism of IMCs of the Sn-xNi/Ni (single-crystal) solder joints (x = 0, 0.05 and 0.1 Wt.%) was investigated.
Keywords
Citation
Wang, J., Chen, J., Zhang, Z., Zhang, P., Yu, Z. and Zhang, S. (2022), "Effects of doping trace Ni element on interfacial behavior of Sn/Ni (polycrystal/single-crystal) joints", Soldering & Surface Mount Technology, Vol. 34 No. 2, pp. 124-133. https://doi.org/10.1108/SSMT-08-2021-0053
Publisher
:Emerald Publishing Limited
Copyright © 2021, Emerald Publishing Limited
The effects of doping minor Ni on the microstructure evolution of an Sn-xNi (x = 0, 0.05 and 0.1 Wt.%)/Ni (Poly-crystal/Single-crystal abbreviated as PC Ni/SC Ni) solder joint during reflow and aging treatment was investigated in this study. Results showed that the intermetallic compounds (IMCs) of the interfacial layer of Sn-xNi/PC Ni joints were Ni3Sn4 phase, while the IMCs of Sn-xNi/SC Ni joints were NiSn4 phase. After the reflow process and thermal aging of different joints, the growth behavior of the interfacial layer was different due to the different mechanisms of element diffusion of the two substrates. The PC Ni substrate mainly provided Ni atoms through grain boundary diffusion. The Ni3Sn4 phase of the Sn0.05Ni/PC Ni joint was finer and the diffusion flux of Sn and Ni elements increased, so the Ni3Sn4 layer of this joint was the thickest. The SC Ni substrate mainly provided Ni atoms through the lattice diffusion. The Sn0.1Ni/SC Ni joint increases the number of Ni atoms at the interface due to the doping of 0.1Ni (Wt.%) elements, so the joint had the thickest NiSn4 layer.
1. Introduction
In the period of electrical packages with a simultaneous increase in the complexity and spatial density, the miniaturization of circuitry has resulted in proportionally smaller volumes of solder at the electrical interconnects. Therefore, there has been a relative increase in the volume fraction of intermetallic compounds (IMCs) in the microstructure of solder joints (Mu et al., 2016). Typically, the Sn-based alloy in a solder joint reacts with the substrate metal during the soldering process and is converted into IMCs (Yang et al., 2020a). It is also well known that the IMC layer acts as the only interconnect medium and its reliability and service performance plays a decisive role and significantly influence, the resultant joint properties (Zhang et al., 2019a). Therefore, the interfacial reactions and IMC growth between an Sn-based solder and the substrate have been of interest to many researchers for a long time.
Copper is the most common conductor metal used in contact with solders, owing to its good solderability characteristics. In an Sn-Cu system, two different types of IMCs (Cu6Sn5 and Cu3Sn) are usually formed at the solder/Cu interface. Meanwhile, a large number of Kirkendall voids (KVs) are often observed at the Cu3Sn/Cu interface, which has detrimental effects on the reliability of joints (Zhang et al., 2019b). Therefore, the interfacial behavior of IMCs, growth, phase evolution and void formation in Sn/Cu systems have attracted much attention (Zhang et al., 2019b; Chen et al., 2019a; Baheti et al., 2017; Wang et al., 2020; Song et al., 2021). It was reported that adding minor alloying elements into the solder reduced the unbalanced diffusion of Cu and Sn, suppressing the formation of the IMC layer and reducing the brittleness of the joint (Yang et al., 2020a; Wang et al., 2020; Song et al., 2021). For example, the grain size of the IMC layer near the solder was refined by Cu, Ag and Ni and it was beneficial in increasing atom diffusion through the Cu6Sn5 layer. This was favorable for the reactions on the Cu6Sn5 side (Yang et al., 2020a). Moreover, the KVs have an impact on the integrity of solder joints for microelectronic applications. It was reported that the formation of voids might be affected by the grain size of the copper substrate (Jian et al., 2010). In recent years, Chen et al. investigated the formation of KVs with three kinds of substrates in Sn/Cu joints during thermal aging. They found that number of voids appeared at the Cu3Sn/Cu interface in Sn/vacuum sputtered Cu and Sn/electroplated Cu joints, but not in Sn/high purity Cu joints (Chen et al., 2019b). Meanwhile, some voids could be mainly caused by the impurities and fine grain incorporated during electroplating. The addition of copper (0.7 Wt.%) retarded the growth of the Cu3Sn layer and suppressed the formation of voids (Chen et al., 2018).
In addition, one of the more note-worthy solder systems was Sn-Ni. In the trend of electronic packaging technology, Ni and Ni-based alloys have become common coating materials in ball grid array microelectronic packaging, owing to the fact that the Ni layer serves as a diffusion barrier layer that can reduce the interfacial reactions between solder alloys and Cu substrates (Zhang et al., 2021; Kim et al., 1999; Chiang et al., 2004). Therefore, the interfacial reactions and IMC growth between Sn-based solder and Ni or Ni-containing substrates have been of interest to many researchers for a long time (Laurila et al., 2006; Li et al., 2015; Yang et al., 2020b; Xu et al., 2019; Osório et al., 2013). For instance, He et al. (2006) reported that the thickness of the IMC increased with an increase in Cu content when soldered with Ni at 260°C and at the same time, the composition and morphology of the IMC had notable changes. A review by Nogita et al. (2012) reported that doped 0.05 Wt.% Ni in Sn-0.7 Wt.% Cu stabilized the high-temperature η-Cu6Sn5 phase down to room temperature and it also included microstructure refinement (Zhao et al., 2007). Wang et al. (2009) found that Ni concentrations higher than 0.01 Wt.% added to an Sn-2.5Ag-0.8Cu alloy could retard the growth of deleterious Cu3Sn particle layers at the alloy/copper interface, even after an aging time of 2,000 h.
However, there have been few studies on the formation of Sn/Ni joints between Sn solder and single-crystal Ni substrates. In addition, it is worth noting that the additions of minor alloying elements to Sn-based lead-free solders have received a lot of attention recently to boost or fine-tune various application properties. Therefore, it is essential to gain better knowledge of the effect of minor Ni additions on the reliability of the Sn-Ni system. As a result, in this work, the effect of doping with Ni on the growth and formation mechanism of IMCs in Sn-xNi/Ni (single-crystal) solder joints (x = 0, 0.05 and 0.1 Wt.%) was investigated.
2. Material and methods
The soldiers used in this work were pure Sn (99.99%), Sn0.05Ni (Wt.%) and Sn0.1Ni (Wt.%). High purity Ni (99.99%) was chosen as the PC Ni substrate. The single-crystal Ni substrate was CMSX-4, which is detailed in ref. Long et al. (2018). Sn solders and Ni substrates were reflowed at 260°C for 5 min to prepare two sets of solder joints and then the joints were cooled at a fast-cooling rate. After reflow, to investigate the interfacial behavior in the solder joints, one set of solder joints was aged isothermally at 180°C for different times (24 h, 48 h, 72 h, 120 h and 168 h). Then the joints, as aged, were mounted in epoxy resin for cross-sectioning followed by surface polishing using grinding papers and then a diamond polishing agent. The cross-section of the interface was observed by scanning electron microscopy (SEM) and the elemental composition of the IMC layer was determined by energy dispersion spectroscopy (EDS). For another set of samples, the top-views of the interface IMC were revealed by etching in an alcohol solution containing 10% HNO3. The interfacial microstructure after reflow was investigated by SEM and the phase analysis of the IMC layer was performed by X-ray diffraction (XRD). In addition, the thickness of the interfacial IMC layer was calculated from the average thickness of each sample using 6–8 SEM images.
3. Results and discussion
3.1 Sn-xNi/Ni (PC/SC) solid-liquid interfacial reaction
Figures 1(a)–1(c) shows cross-sectional SEM micrographs of the Sn/polycrystalline Ni, Sn0.05Ni/polycrystalline Ni and Sn0.1Ni/polycrystalline Ni samples after reflow, while 1d and 1e show enlarged images of the marked interface areas B1 and C1, respectively. The IMC layer formed was analyzed by EDS. The results demonstrated that the ratio of Ni and Sn (in at.%) of the three joints was 3:4, corresponding to the Ni3Sn4 phase. It can be found that the IMC layer becomes substantially thicker after adding elemental nickel into the solder, in contrast to that of pure Sn solder. Moreover, the thickest IMC layer was observed when the Ni content was 0.05 Wt.% in the solder. Besides, the microstructure of the IMC layer of Sn0.05Ni and Sn0.1Ni solders were compared, as shown in Figures 1(d) and 1(e). It can be found that the separated IMC particles congregated together, forming a whole bulky IMC layer when the Ni content increased.
Figures 2(a)–2(c) show the morphology of IMCs in the solder joints of Sn-xNi/SC Ni with Ni contents of 0, 0.05 and 1 Wt.%, respectively. It was found that the ratio of Ni and Sn (in at.%) of the IMC formed by the three joints was 1:4, corresponding to the NiSn4 phase. It was also found, that after reflow at 260°C, the NiSn4 layer of pure Sn solder was clearly visible and appeared homogeneous [Figure 2(a)]. More Ni content in the solder resulted in a thicker IMC layer, although the difference was not obvious. For the Sn0.1Ni solder, the interface layer was the thickest. In addition, comparing Figure 1, the NiSn4 layer of the joints using single-crystal Ni as the substrate was smaller than the thickness of the Ni3Sn4 layer of the joints using polycrystalline Ni as the substrate.
For the sake of better understanding the IMC growth behavior, the interfacial IMC grains were further researched. Figures 3(a)–3(c) display the top views of the Sn-xNi/PC Ni solder system after reflow. As can be observed, the Ni3Sn4 grains were dispersed homogeneously and compacted, as well as scallop-shaped. Two Ni3Sn4 morphologies appeared at the interface of the Sn/PC Ni solder joints: scallop-shaped and faceted-polygonal [Figure 3(a)]. Compared with the Ni3Sn4 of pure Sn solder, it can be found that Ni3Sn4 grains were much finer as Sn solder was doped with nickel. Notably, the scallop-shaped Ni3Sn4 grains of Sn0.05Ni/PC Ni joint were the finest.
Figures 4(a)–4(c) shows the top views of the IMC layers of Sn-xNi/SC Ni with Ni contents of 0, 0.05 and 1.00 Wt.%, respectively. According to Figure 4, the interfacial NiSn4 crystals had a faceted, tile-like morphology. By comparing the SEM images, some microstructural differences among the IMCs of these three kinds of solders can be identified. As the Ni content increased, the (001) facet area became larger whereas their thickness increased slightly.
For each sample, the excess Sn solder was removed and the phase of the IMC layer was analyzed via the XRD pattern. The XRD pattern analysis of the IMC layer in Figure 5(a) shows that the IMC layer with polycrystalline Ni as the substrate was the Ni3Sn4 phase. Figure 5(b) shows the XRD pattern of the IMCs layer with single-crystal Ni as the substrate (Yang et al., 2020b). By comparing the ideal structure of PtSn4 and β-IrSn4, it was found that the diffraction pattern tested by XRD showed a narrow Bragg reflection and a characteristic wide diffraction band indicating stacking disorder in the NiSn4 crystal structure. It is speculated that the IMCs of the Sn/SC Ni joint were in the NiSn4 phase (Schimpf et al., 2016).
3.2 Sn-xNi/Ni (PC/SC) solid-solid interfacial reaction
Thermal aging testing is generally used to accelerate the growth of IMCs in an elevated temperature environment. In this study, the aging temperature was 180°C and the test was continued for 168 h. The IMC thicknesses after aging for 120 h and 168 h were compared, as shown in Figure 6. It can be distinctly observed that the thickness of the Ni3Sn4 layer continued to increase with thermal aging time and the thickness of the IMC layer of the Sn0.05Ni/PC Ni joint was the thickest. Besides, as the aging time extended, the phase composition of the IMC layers remained unchanged.
The samples with single-crystal Ni as the substrate were thermally aged and the phase composition of the IMC layer also remained unchanged. Figures 7(a)–7(f) show cross-sectional SEM images of the interfaces formed between single-crystal Ni and Sn, Sn0.05Ni and Sn0.1Ni solder joints aged for 120 h and 168 h at 180°C, respectively. With the extension of the thermal aging time, the thickness of the NiSn4 layer continued to increase and the thickness of the NiSn4 layer of the Sn0.1Ni/SC Ni joint was the thickest. When the reflow samples were solid-state aged at 180°C for 168 h the interfacial NiSn4 layer became rougher. Compared with the Ni3Sn4 layer of the sample with polycrystalline Ni as the substrate, the thicknesses of the NiSn4 layers of different joints varied greatly and the growth of the NiSn4 layer was uneven.
To understand the growth kinetics of Ni3Sn4 and NiSn4, Figure 6 shows the average thickness of the IMC layers of different Sn/Ni joints at different aging times. Figure 6(a) shows the thickness variation of the Ni3Sn4 layer of different joints. It was found that the thickness of the Ni3Sn4 layer of the Sn0.05Ni/PC Ni joint was the largest. The thickness of the Ni3Sn4 layer of Sn0.1Ni/PC Ni joint was not much different from that of Sn0.05Ni/PC Ni joint, but the thickness of the Ni3Sn4 layer of these two joints was larger than that of the Sn/PC Ni joint. Figure 6(c) shows the thickness variation of NiSn4 layers of different joints. It was found that the thickness of the NiSn4 layer of Sn0.1Ni/SC Ni was the thickest. The thicknesses of the NiSn4 layer of Sn0.1Ni/SC Ni, Sn0.05Ni/SC Ni and Sn/SC Ni joints were quite different and the thickness of the NiSn4 layer of Sn/SC Ni joints was the smallest. Comparing Figures 6(a) and 6(c), for the joints with polycrystalline Ni and single-crystal Ni as the substrate, the Sn solder doped with trace amounts of Ni could promote the growth of Ni3Sn4 layer and NiSn4 layer, where the thickness of NiSn4 layer was significantly thinner than that of Ni3Sn4 layer thickness.
Due to the different substrates, the growth mechanisms of the IMC layers of the joints were different. The relationship between ln(d-d0) and ln(t) of the Ni3Sn4 layer and NiSn4 layer of different joints is shown in Figures 6(b) and 6(d). The diffusion coefficient (D) and time exponent (n) can be calculated and the data is listed in Table 1 (Figure 8).
The thickness of the IMC layers changed with aging time and can be expressed by the following empirical power law (Kumar and Chen, 2011):
Equation (1) can be converted into:
The time exponent, n, can predict the control mechanism of the IMC growth. If n is 1, the IMC growth is controlled by the reaction rate. As n is 2, the growth mechanism is controlled by diffusion of the reaction elements (Kumar and Chen, 2011; Liu et al., 2013). In this work, the n values of Sn-xNi/SC Ni (x = 0, 0.05 and 0.1) joints were all close to 1, indicating that the growth of the NiSn4 layer was controlled by the reaction rate. The n values of Sn-xNi/PC Ni (x = 0, 0.05 and 0.1) joints were all close to 2, indicating that the growth of the Ni3Sn4 layer was mainly controlled by diffusion.
3.3 Discussion
Figure 9 gives a microstructural and schematic illustration of the polycrystalline Ni and single-crystal nickel substrates. The figure shows that the polycrystalline Ni substrate had a large number of grain boundaries, as shown in Figures 9(a) and 9(c). For Sn-xNi/polycrystalline Ni joints, nickel atoms formed Ni3Sn4 phases with Sn atoms mainly through grain boundary diffusion. The reaction formula was as follows:
The greater the number of grain boundaries, the greater the diffusion flux of the Ni atoms and the faster the growth rate of the Ni3Sn4 layer. Refer to the top-view of the Ni3Sn4 layer after reflowing, as shown in Figures 3(a), 3(b) and 3(c). It was shown that the Ni3Sn4 phase had a highly irregular morphology with diverging and converging scallops because monoclinic Ni3Sn4 grows only along [010] without a reproducible orientation relationship with Sn (Mita et al., 2005). With an increase in Ni content, the number and size of the Ni3Sn4 clusters in the melt increased. These clusters acted as heterogeneous nucleation sites at the interface to promote the nucleation of IMCs, thereby making the initial grain size of Ni3Sn4 decrease and increase in number. Therefore, the Ni3Sn4 grains of Sn0.05Ni/polycrystalline Ni and Sn0.1Ni/polycrystalline Ni were smaller than those of Sn/polycrystalline Ni joints. As Figures 3(a) and 3(b) show, comparing Sn0.1Ni/polycrystalline Ni and Sn0.05Ni/polycrystalline Ni joints, the content of Ni in the Sn solder was higher, which made the concentration gradient of elemental Ni at the interface smaller, resulting in the diffusion flux of Ni being relatively small. A large number of Ni atoms diffused from the grain boundaries and tin atoms preferentially nucleated on the Ni3Sn4 clusters. The Sn0.1Ni/polycrystalline Ni joint had a large number of Ni3Sn4 clusters, which made the Ni3Sn4 phase formed by the combination of adjacent Ni3Sn4 clusters grow further, resulting in the grain size of the Ni3Sn4 formed by the Sn0.1Ni/polycrystalline Ni greater joint than that of the Sn0.05Ni/polycrystalline Ni joint and the number of grain boundaries was less than that of the Sn0.05Ni/polycrystalline Ni joint. Therefore, the Ni3Sn4 phase of the Sn0.05Ni/polycrystalline Ni joint had the smallest grain size and the largest number of grain boundaries, which could provide the most Ni atoms at the interface and the diffusion flux of the nickel atoms was the largest. Thus, the thickness of the Ni3Sn4 layer was the thickest after aging.
The microstructure of the single crystal Ni substrate consisted of a single-crystal Ni substrate (denoted as γ) and a strengthened Ni3Al precipitate (denoted as γ’) (Shishvan et al., 2017). As the γ’, phase was precipitated in the γ’ matrix phase in a coherent form, the single crystal Ni substrate had almost no grain boundaries, as shown in Figures 7(b) and 7(d). For Sn-xNi/single-crystal Ni (x = 0, 0.05 and 0.1) joints, Ni atoms formed a NiSn4 phase with Sn atoms mainly through lattice diffusion. The Ni atom diffusion flux of lattice diffusion was much smaller than that of grain boundary diffusion. The number of Ni atoms provided by Sn/single-crystal Ni joints through lattice diffusion was small, thus the NiSn4 phase was formed at the interface, as shown in Figure 2. The reaction formula is as follows:
This is also the reason that the reflowed Ni3Sn4 layer on polycrystalline Ni was thicker than the NiSn4 layer formed on single crystalline Ni. The Sn0.05Ni/single-crystal Ni and Sn0.1Ni/single-crystal Ni joints contained a small amount of nickel in the tin solder, compared to the Sn/single-crystal Ni joint, which could provide more nickel atoms at the joint interface and form more NiSn4 clusters at the interface. With the increase in nickel doping, more NiSn4 clusters could be formed at the interface of the Sn0.1Ni/single-crystal Ni joints than the Sn0.05Ni/single-crystal Ni joints, then the NiSn4 clusters at the interface combined with the adjacent NiSn4 to grow. This caused the NiSn4 phase at the interface to continuously grow and become thicker, resulting in the number of NiSn4 phases in the Sn0.1Ni/single-crystal Ni joint being smaller and the thickness greater. As the number of NiSn4 phases at the interface of Sn0.1Ni/single-crystal Ni joints was less, the diffusion flux of Ni atoms increased, so the interfacial NiSn4 layer of the Sn0.1Ni/single-crystal Ni joints was the thickest among the three joints.
4. Conclusions
In this study, the effects of doping with minor amounts of Ni on the microstructural evolution of an Sn-xNi (x = 0, 0.05 and 0.1 Wt.%)/Ni (poly-crystalline/single-crystal) solder joints during reflow and aging treatment have been investigated. The results are summarized as follows:
The polycrystalline Ni substrate provided Ni atoms through grain boundary diffusion, whereas for the single-crystal Ni substrate it was through lattice diffusion. The polycrystalline Ni substrate and the single-crystal Ni substrates provided different diffusion fluxes of Ni atoms at the interface, resulting in different phases of IMCs. IMCs of the interfacial layer of Sn-xNi/polycrystalline Ni joints were a scallop-shaped Ni3Sn4 phase, while the IMCs of Sn-xNi/single-crystal Ni joints were a faceted tile-like NiSn4 phase.
With the addition of nickel, the interface phase was not changed. Meanwhile, doping with trace amounts of nickel refined the grains, increased the diffusion channels of Ni atoms, then increased the diffusion of Ni and Sn and ultimately promoted the growth of the IMC. The Ni3Sn4 phase of the Sn0.05Ni/PC Ni joint had the smallest grain size and the largest number of grain boundaries, which could provide the most nickel atoms at the interface and the diffusion flux of nickel atoms was the largest. Thus, the thickness of the Ni3Sn4 layer was the largest. As for single crystal Ni, with the increase in nickel doping, the number of NiSn4 phases at the interface of the Sn0.1Ni/single-crystal Ni joints was less, the diffusion flux of the nickel atoms increased, so the interfacial NiSn4 layer of Sn0.1Ni/single-crystal Ni joints was the thickest.
Due to the different substrates, the growth mechanisms of the IMC layers of the joints were different. The n values of the Sn-xNi/single-crystal Ni (x = 0, 0.05 and 0.1) joints were all close to 1, indicating that the growth of the NiSn4 layer was controlled by the reaction rate. The n values of the Sn-xNi/polycrystalline Ni (x = 0, 0.05 and 0.1) joints were all close to 2, indicating that the growth of the Ni3Sn4 layer was mainly controlled by elemental diffusion.
Figures
Figure 5
XRD patterns for interfacial IMCs of solder joints. Ni3Sn4 layer (a) and NiSn4 layer (b) Yang et al. (2020b)
Kinetic parameters for different joints at an aging temperature of 180°C
| Solder joints | n | K (μm/h1/n) | D (m–17/s) | Solder joints | n | K (μm/h1/n) | D (m–12/s) |
|---|---|---|---|---|---|---|---|
| Sn/PC Ni | 2.008 | 0.407 | 4.6 | Sn/SC Ni | 0.995 | 0.021 | 5.8 |
| Sn0.05Ni/PC Ni | 2.174 | 0.52 | 7.5 | Sn0.05Ni/SC Ni | 1.284 | 0.09 | 25 |
| Sn0.1Ni/PC Ni | 2.151 | 0.493 | 6.8 | Sn0.1Ni/SC Ni | 1.292 | 0.105 | 29.2 |
References
Baheti, V.A., Kashyap, S., Kumar, P., Chattopadhyay, K. and Paul, A. (2017), “Bifurcation of the Kirkendall marker plane and the role of Ni and other impurities on the growth of Kirkendall voids in the Cu-Sn system”, Acta Materialia, Vol. 131, pp. 260-270.
Chen, J., Yang, J., Zhang, Y., Yu, Z. and Zhang, P. (2019b), “Effect of substrates on the formation of Kirkendall voids in Sn/Cu joints”, Welding in the World, Vol. 63 No. 3, pp. 751-757.
Chen, J., Zhang, Y., Yu, Z., Zhao, W. and Wu, D. (2018), “Interface growth and void formation in Sn/Cu and Sn0.7Cu/Cu systems”, Applied Sciences, Vol. 8 No. 12, p. 2703.
Chen, J., Zhang, K., Zhang, P., Yu, Z., Zhang, Y., Yu, C. and Hao, L. (2019a), “The Zn accumulation behavior, phase evolution and void formation in Sn-xZn/Cu systems by considering trace Zn: a combined experimental and theoretical study – science direct”, Journal of Materials Research and Technology, Vol. 8 No. 5, pp. 4141-4150.
Chiang, M.J., Chang, S.Y. and Chuang, T.H. (2004), “Reflow and burn-in of a Sn-20In-0.8Cu ball grid array package with a Au/Ni/Cu pad”, Journal of Electronic Materials, Vol. 33 No. 1, pp. 34-39.
He, D.P., Yu, D.Q., Wang, L. and Wu, C. (2006), “Effect of cu content on IMC between Sn-Cu solder and Cu and Ni substrates”, The Chinese Journal of Nonferrous Metals, Vol. 16 No. 4, pp. 701-708.
Jian, Z., Mo, L., Wu, F., Bo, W. and Wu, Y. (2010), Effect of Cu Substrate and Solder Alloy on the Formation of Kirkendall Voids in the Solder Joints during Thermal Aging, IEEE, pp. 944-948.
Kim, P.G., Jang, J.W., Lee, T.Y. and Tu, K.N. (1999), “Interfacial reaction and wetting behavior in eutectic Sn/Pb solder on Ni/Ti thin films and Ni foils”, Journal of Applied Physics, Vol. 86 No. 12, pp. 6746-6751.
Kumar, A. and Chen, Z. (2011), “Interdependent intermetallic compound growth in an electroless Ni-P/Sn-3.5Ag reaction couple”, Journal of Electronic Materials, Vol. 40 No. 2, pp. 213-223.
Laurila, T., Vuorinen, V. and Kivilahti, J.K. (2006), “Interfacial reactions between lead-free solders and common base materials”, Materials Science and Engineering: R: Reports, Vol. 49 Nos 1/2, pp. 1-60.
Li, Y., Zhang, Y., Dai, J., Jing, Y., Ge, J. and Ning, Z. (2015), “Microstructure, interfacial IMC and mechanical properties of Sn–0.7Cu–xAl (x = 0-0.075) lead-free solder alloy”, Materials & Design, Vol. 67, pp. 209-216.
Liu, X., Tian, Y.E., Wu, Y.P., Sitaraman, S.K. and Chow, J. (2013), “Study of intermetallic growth and kinetics in fine-pitch lead-free solder bumps for next-generation flip-chip assemblies”, Journal of Electronic Materials, Vol. 42 No. 2, pp. 230-239.
Long, H., Mao, S., Liu, Y., Zhang, Z. and Han, X. (2018), “Microstructural and compositional design of Ni-based single crystalline superalloys ― a review”, Journal of Alloys and Compounds, Vol. 743, pp. 203-220.
Mita, M., Kajihara, M., Kurokawa, N. and Sakamoto, K. (2005), “Growth behavior of Ni3Sn4 layer during reactive diffusion between Ni and Sn at solid-state temperatures”, Materials Science and Engineering: A, Vol. 403 Nos 1/2, pp. 269-275.
Mu, D.K., Mcdonald, S.D., Read, J., Huang, H. and Nogita, K. (2016), “Cheminform abstract: critical properties of Cu6Sn5 in electronic devices: recent progress and a review”, Current Opinion in Solid State & Materials Science, Vol. 47 No. 19, pp. 55-76.
Nogita, K., Mu, D., Mcdonald, S.D., Read, J. and Wu, Y.Q. (2012), “Effect of Ni on phase stability and thermal expansion of Cu6xNixSn5 (x = 0, 0.5, 1, 1.5 and 2)”, Intermetallics, (none), Vol. 26, pp. 78-85.
Osório, W.R., Leiva, D.R., Peixoto, L.C., Garcia, L.R. and Garcia, A. (2013), “Mechanical properties of Sn-Ag lead-free solder alloys based on the dendritic array and Ag3Sn morphology”, Journal of Alloys and Compounds, Vol. 562, pp. 194-204.
Schimpf, C., Kalanke, P., Shang, S.L., Liu, Z.K. and Leineweber, A. (2016), “Stacking disorder in metastable nisn4”, Materials & Design, Vol. 109 No. 5, pp. 324-333.
Shishvan, S.S., Mcmeeking, R.M., Pollock, T.M. and Deshpande, V.S. (2017), “Discrete dislocation plasticity analysis of the effect of interfacial diffusion on the creep response of Ni single-crystal superalloys”, Acta Materialia, Vol. 135, pp. 188-200.
Song, R.W., Fleshman, C.J., Wang, Y.C., Tsai, S.Y. and Duh, J.G. (2021), “IMC suppression and phase stabilization of Cu/Sn-Bi/Cu microbump via Zn doping”, Materials Letters, Vol. 282 No. 2, p. 128735.
Wang, Q., Zhang, S., Liu, G., Lin, T. and He, P. (2020), “The mixture of silver nanowires and nanosilver-coated copper micron flakes for electrically conductive adhesives to achieve high electrical conductivity with low percolation threshold”, Journal of Alloys and Compounds, Vol. 820, p. 153184.
Wang, Y.W., Chang, C.C. and Kao, C.R. (2009), “Minimum effective Ni addition to SnAgCu solders for retarding Cu3Sn growth”, Journal of Alloys and Compounds, Vol. 478 Nos 1/2, pp. L1-L4.
Xu, S., Qi, X., Xu, X., Wang, X., Yang, Z., Zhang, S., Lin, T. and He, P. (2019), “Effects of electroless nickel plating method for low temperature joining ZnS ceramics”, Journal of Materials Science: Materials in Electronics, Vol. 30 No. 16, pp. 15236-15249.
Yang, M., Chen, J., Yang, J., Zhang, P., Yu, Z. and Zeng, Z. (2020a), “Interfacial transfer and phase evolution between Cu and Sn solder doped with minor Cu, Ag and Ni: experimental and theoretical investigations”, Applied Physics A, Vol. 126 No. 8, pp. 1-12.
Yang, M.Y., Chen, J.S., Yang, J., Zhang, P.L. and Lu, H. (2020b), “Effect of crystal structure of nickel substrates on interfacial behaviors in Sn/Ni soldered joints”, Materials Letters, Vol. 278, p. 128424.
Zhang, S., Wang, Q., Lin, T., Zhang, P., He, P. and Paik, K.W. (2021), “Cu-Cu joining using citrate coated ultra-small nano-silver paste”, Journal of Manufacturing Processes, Vol. 62, pp. 546-554.
Zhang, H.K., Chen, J.S., Zhang, L.X., Yu, Z., Zhang, P., Zhang, Y.Z., Yu, C. and Lu, H. (2019a), “Phase stability, elasticity, hardness and electronic structures for binary MnBm (M = Ni, Cr, Mo, W, n = 23, 5, 3, 1, m = 6, 3, 2, 1) borides: a comprehensive study using first principles”, Phase Transitions, Vol. 93 No. 1, pp. 158-174.
Zhang, S., Xu, X., Lin, T. and He, P. (2019b), “Recent advances in nano-materials for packaging of electronic devices”, Journal of Materials Science: Materials in Electronics, Vol. 30 No. 15, pp. 13855-13868.
Zhao, N., Ma, H.T. and Wang, L. (2007), “Interfacial reactions between Sn-Cu based multicomponent solders and Ni substrates during soldering and aging”, International Symposium on High Density Packaging & Microsystem Integration, IEEE.
Acknowledgements
This project is supported by National Natural Science Foundation of China (Grant No. 51805316), China postdoctoral Science Foundation (No.2019M651491), State Key Laboratory of Advanced Welding and Joining (No. AWJ-20-M12), Shanghai Local Universities Capacity Building Project of Science and Technology Innovation Action Program (20030500900).